• 한국멀티미디어학회논문지
• /
• 제19권6호
• /
• pp.1024-1033
• /
• 2016

In this study, a system is proposed for analyzing whether driver's eyes are open or closed on the basis of images to determine driver's drowsiness. The proposed system converts eye areas detected by a camera to a color space area to effectively detect eyes in a dark situation, for example, tunnels, and a bright situation due to a backlight. In addition, the system used a thickness distribution of a detected eye area as a feature value to analyze whether eyes are open or closed through the Support Vector Machine(SVM), representing 90.09% of accuracy. In the experiment for the images of driver wearing glasses, 83.83% of accuracy was obtained. In addition, in a comparative experiment with the existing PCA method by using Eigen-eye and Pupil Measuring System the detection rate is shown improved. After the experiment, driver's drowsiness was identified accurately by using the method of summing up the state of driver's eyes open and closes over time and the method of detecting driver's eyes that continue to be closed to examine drowsy driving.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제15권3호
• /
• pp.186-191
• /
• 2015

Hand gesture is the most common tool used to interact with and control various electronic devices. In this paper, we propose a novel hand gesture recognition method using fuzzy logic based classification with a new type of sensor array. In some cases, feature patterns of hand gesture signals cannot be uniquely distinguished and recognized when people perform the same gesture in different ways. Moreover, differences in the hand shape and skeletal articulation of the arm influence to the process. Manifold features were extracted, and efficient features, which make gestures distinguishable, were selected. However, there exist similar feature patterns across different hand gestures, and fuzzy logic is applied to classify them. Fuzzy rules are defined based on the many feature patterns of the input signal. An adaptive neural fuzzy inference system was used to generate fuzzy rules automatically for classifying hand gestures using low number of feature patterns as input. In addition, emotion expression was conducted after the hand gesture recognition for resultant human-robot interaction. Our proposed method was tested with many hand gesture datasets and validated with different evaluation metrics. Experimental results show that our method detects more hand gestures as compared to the other existing methods with robust hand gesture recognition and corresponding emotion expressions, in real time.

• 국제초고층학회논문집
• /
• 제11권2호
• /
• pp.115-124
• /
• 2022

Welding is a significant part of the construction industry. Since most high-rise building construction structures rely on a robust metal frame welded together, welding defect can damage welded structures and is critical to safety and quality. Despite its importance and heavy usage in construction, the labor shortage of welders has been a continuous challenge to the construction industry. To deal with the labor shortage, the ultimate goal of this study is to design and develop an automated robotic welding system composed of a welding machine, unmanned ground vehicle (UGV), robotic arm, and visual sensors. This paper proposes and focuses on automated weaving using the robotic arm. For automated welding operation, a microcontroller is used to control the switch and is added to a welding torch by physically modifying the hardware. Varying weave patterns are mathematically programmed. The automated weaving is tested using a brush pen and a ballpoint pen to clearly see the patterns and detect any changes in vertical forces by the arm during weaving. The results show that the weave patterns have sufficiently high consistency and precision to be used in the actual welding. Lastly, actual welding was performed, and the results are presented.

• 한국지능시스템학회논문지
• /
• 제23권6호
• /
• pp.551-557
• /
• 2013

지능형 로봇 연구 분야에 있어, 손을 이용한 제스처 인식은 매우 중요한 연구 분야로 간주 되고 있으며, 스마트 폰, 스마트 TV 등에 상용화 되어왔다. 제스처 인식에 있어, 강인한 손 검출 기술을 필수적인데, 손의 모양이 일정치 않고, 복잡한 배경이나 조명변화 아래서는 손 검출이 쉽지 않다는 어려움이 있다. 본 논문은 실내 환경에서 사용자가 가리키는 방향을 인식하기 위한 손 검출 알고리즘을 제안한다. 손 검출에 대한 오검출을 최대한 줄이기 위해, 머리-어깨 검출 결과를 기반으로 손 검색 영역을 한정시키고, 피부색을 이용해 최소한의 후보군들을 발생시켜, HOG-SVM을 이용하여 손을 검출하였다. 그리고 머리-어깨, 손 검출 결과를 통해 팔의 방향 각도를 추정하였다. 제안된 방법은 실제 실내 환경에서 추출된 영상을 통해 실험을 진행하였고, 강인한 성능을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권11호
• /
• pp.4084-4104
• /
• 2021

Visual place recognition (VPR) is a fundamental yet challenging task of mobile robot navigation and localization. The existing VPR methods are usually based on some pairwise similarity of image descriptors, so they are sensitive to visual appearance change and also computationally expensive. This paper proposes a simple yet effective four-step method that achieves adaptive convolutional neural network (CNN) landmark matching for VPR. First, based on the features extracted from existing CNN models, the regions with higher significance scores are selected as landmarks. Then, according to the coordinate positions of potential landmarks, landmark matching is improved by removing mismatched landmark pairs. Finally, considering the significance scores obtained in the first step, robust image retrieval is performed based on adaptive landmark matching, and it gives more weight to the landmark matching pairs with higher significance scores. To verify the efficiency and robustness of the proposed method, evaluations are conducted on standard benchmark datasets. The experimental results indicate that the proposed method reduces the feature representation space of place images by more than 75% with negligible loss in recognition precision. Also, it achieves a fast matching speed in similarity calculation, satisfying the real-time requirement.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권10호
• /
• pp.836-850
• /
• 2009

포즈인식은 최근에 유비쿼터스 환경, 행위 예술, 로봇 제어 등에서 그 필요성이 증가되고 있는 분야로써, 컴퓨터비전, 패턴인식 등에서 활발히 연구되고 있다. 하지만 기존의 포즈인식 연구들은 사람의 회전이나 이동에 따라서 불안정한 인식률을 보인다는 단점을 갖고 있다. 이는 포즈 인식을 위해 추출한 특징이 사람의 회전, 이동 등의 다양한 변수에 영향을 크게 받기 때문이다. 이를 극복하기 위하여 본 논문에서는, 다 시점(multi-view) 환경에서의 3D Star Skeleton과 주성분 분석(principal component analysis: PCA)에 기반한 사람의 회전에 강건한 특징 추출을 제안한다. 제안된 시스템은 포즈의 특징 추출을 위해 다 시점 환경 기반의 visual hull을 생성하는 과정에서 획득 가능한 깊이 정보를 표현하는 8개의 projection map을 입력데이터로 사용한다. 이를 통해 포즈의 3D 정보를 반영하는 3D Star Skeleton을 구성하고 주성분 분석 기반의 회전에 강건한 특징을 추출한다. 실험결과에서는 다양하게 회전된 사람으로부터 생성된 3D Star Skeleton에서 특징을 추출하고 다양한 인식기를 통해 포즈인식을 해보았으며, 제안된 특징 추출 방법이 사람의 회전에 강건함을 알 수 있었다.